The goal of this proposal is to provide a program of mentoring, clinical and didactic education, and a research plan that will allow Dr. Howard Rosen to supplement his skills in behavioral neurology and neuroimaging with training in theoretical and experimental models of emotional processing as well as statistical analysis. The particular focus of this proposal is based on the underlying principle that the study of emotional abnormalities in patients with diseases causing dementia will broaden our understanding of these diseases, improve our ability to differentiate them from each other and yield valuable insights into the neuroanatomic basis of emotional processing. Alzheimer's disease (AD) and frontotemporal lobar degeneration (FTLD), the two most common forms of neurodegenerative dementia, are each associated with distinct patterns of behavioral abnormalities. In particular, the diagnosis of FTLD requires the recognition of behavioral signs whose characteristics strongly suggest disorders of emotion. Yet, very few studies have directly assessed emotional processing in AD and FTLD. The proposed research project will compare emotional processing in 46 patients with AD, 46 patients with FTLD, and 46 age-matched controls. Two major aspects of emotional processing will be examined. Emotional understanding (the ability to appreciate emotion in others) will be assessed using bedside measures, and emotional reactivity (the ability to develop appropriate responses to emotional situations) will be studied during the elicitation of emotion with films presented on video in a laboratory setting. Emotional reactivity will be measured via participants' self-report, monitoring of autonomic function (galvanic skin response, heart rate, blood pressure, etc.) and monitoring of behavior (facial actions and somatic movements). The specific hypotheses regarding the experimental groups are 1) the frontal and temporal variants of FTLD (fvFTLD and tvFTLD, respectively) will be associated with deficits in emotional understanding when compared with controls and AD and 2) fvFTLD and tvFTLD will be associated with deficits in emotional reactivity compared with controls and AD. In addition, MRI scans will also be obtained for measurements of frontal cortical and amygdala volumes in order to test two more hypotheses: 1) emotional understanding and reactivity will be correlated with fight amygdala volume in tvFTLD and 2) emotional understanding and reactivity will be correlated with fight frontal cortical volume in fvFTLD.